The primary aim of this paper is to introduce a new class of functions to facilitate the design of cryptographically good functions. It turns out that
|
|
- Ella Cummings
- 5 years ago
- Views:
Transcription
1 Plateaued Functions Yuliang Zheng 1 and Xian-Mo Zhang 1 Monash University, Frankston, Melbourne, VIC 3199, Australia yuliang.zheng@monash.edu.au, The University ofwollongong, Wollongong, NSW 5, Australia xianmo@cs.uow.edu.au Abstract. The focus of this paper is on nonlinear characteristics of cryptographic Boolean functions. First, we introduce the notion of plateaued functions that have many cryptographically desirable properties. Second, we establish a sequence of strengthened inequalities on some of the most important nonlinearity criteria, including nonlinearity, propagation and correlation immunity, and prove that critical cases of the inequalities coincide with characterizations of plateaued functions. We then proceed to prove that plateaued functions include as a proper subset all partiallybent functions that were introduced earlier by Carlet. This settles an open question that arises from previously known results on partiallybent functions. In addition, we construct plateaued, but not partiallybent, functions that have many properties useful in cryptography. Key Words Bent Functions, Cryptography, Nonlinear Characteristics, Partially-Bent Functions, Plateaued Functions. 1 Motivations In the design of cryptographic functions, one often faces the problem of ful- lling the requirements of a multiple number of nonlinearity criteria. Some of the requirements contradict others. The most notable example is perhaps bent functions while these functions achieve the highest possible nonlinearity and satisfy the propagation criterion with respect to every non-zero vector, they are not balanced, not correlation immune and exist only when the number of variables is even. Another example that clearly demonstrates how some nonlinear characteristics may impede others is partially-bent functions introduced in []. These functions include bent functions as a proper subset. Partially-bent functions are interesting in that they can be balanced and also highly nonlinear. However, except those that are bent, all partially-bent functions have non-zero linear structures, which are considered to be cryptographically undesirable.
2 The primary aim of this paper is to introduce a new class of functions to facilitate the design of cryptographically good functions. It turns out that these cryptographically good functions maintain all the desirable properties of partiallybent functions while possess no non-zero liner structures. This class of functions are called plateaued functions. To study the properties of plateaued functions, we establish a sequence of inequalities concerning nonlinear characteristics of functions. We show that plateaued functions can be characterized by the critical cases of these inequalities. In particular, we demonstrate that plateaued functions reach the upper bound on nonlinearity given by the inequalities. We also examine relationships between plateaued functions and partiallybent functions. We show that partially-bent functions must be plateaued while that the converse is not true. This disproves a conjecture and motivates us to construct plateaued functions without non-zero linear structures. Other useful properties of plateaued functions include that they exist for both even and odd numbers of variables, can be balanced and correlation immune. Boolean Functions Denition 1. We consider functions from V n to GF () (or simply functions on V n ), V n is the vector space ofn tuples of elements from GF (). Usually we write a function f on V n as f(x), where x =(x 1 ::: x n ) is the variable vector in V n. The truth table of a function f on V n is a (0 1)-sequence dened by (f( 0 ) f( 1 ) ::: f( )), and the sequence of f is a (1 ;1)-sequence de- ned by((;1) f (0) (;1) f (1) ::: (;1) f (n;1) ),where 0 =(0 ::: 0 0), 1 = (0 ::: 0 1), :::, n;1 ;1 =(1 ::: 1 1). Thematrix of f is a (1 ;1)-matrix of order n dened bym =((;1) f (ij ) ) where denotes the addition in GF (). f is said to be balanced if its truth table contains an equal number of ones and zeros. Given two sequences ~a =(a 1 a m )and ~ b =(b 1 b m ), their componentwise product is dened by~a~ b =(a1 b 1 a m b m )andthescalar product of ~a and ~ b, denoted by h~a ~ bi, is dened as the sum of the component-wise multiplications, where the operations are dened in the underlying eld. In particular, if m = n and ~a, ~ b are the sequences of functions f and g on V n respectively, then~a ~ b is the sequence of f g where denotes the addition in GF (). An ane function f on V n is a function that takes the form of f(x 1 ::: x n )= a 1 x 1 a n x n c, where denotes the addition in GF () and a j c GF (), j =1 ::: n.furthermore f is called a linear function if c =0. A(1 ;1)-matrix A of order m is called a Hadamard matrix if AA T = mi m, where A T is the transpose of A and I m is the identity matrix of order m. A Sylvester-Hadamard matrix of order n, denoted by H n, is generated by the following recursive relation H 0 =1 H n = Hn;1 H n;1 H n;1 ;H n;1 n =1 ::::
3 Let `i, 0 i n ; 1, be the i row ofh n.then`i is the sequence of a linear function ' i (x) dened by the scalar product ' i (x) =h i xi, where i V n is the binary representation of integer i, i =0 1 ::: n ; 1. The Hamming weight of a (0 1)-sequence, denoted by W (), is the number of ones in the sequence. Given two functions f and g on V n, the Hamming distance d(f g) between them is dened as the Hamming weight of the truth table of f(x) g(x), where x =(x 1 ::: x n ). Denition. The nonlinearity of a function f on V n, denoted by N f, is the minimal Hamming distance between f and all ane functions on V n, i.e., N f = min i=1 ::: n+1 d(f ' i ) where ' 1, ', :::, ' n+1 are allthe ane functions on V n. The following characterizations of nonlinearity will be useful (for a proof see for instance [6]). Lemma 1. The nonlinearity of f can be expressed by N f = n;1 ; 1 maxfjh `iij 0 i n ; 1g where is the sequence off and `i is the ith row of H n, i =0 1 ::: n ; 1. Denition 3. Let f be a function on V n. For a vector V n, denote by () the sequence off(x ). Thus (0) is the sequence off itself and (0) () is the sequence off(x) f(x ). Set () =h(0) ()i, the scalar product of (0) and (). () is also called the auto-correlation of f with a shift. Denition 4. Let f be a function on V n.wesaythatf satises the propagation criterion with respect to if f(x) f(x ) is a balanced function, where x =(x 1 ::: x n ) and is a vector in V n. Furthermore f is said to satisfy the propagation criterion of degree k if it satises the propagation criterion with respect to every non-zero vector whose Hamming weight is not larger than k (see [7]). The strict avalanche criterion (SAC) [11] is the same as the propagation criterion of degree one. Obviously, () = 0 if and only if f(x)f(x) is balanced, i.e., f satises the propagation criterion with respect to. Denition 5. Let f be a function on V n. V n is called alinear structure of f if j()j = n. For any function f, ( 0 )= n,where 0 = 0, the zero vector on V n. Hence the zero vector is a linear structure of every function on V n.itiseasytoverify that the set of all linear structures of a function f form a subspace of V n,whose dimension is called the linearity of f. It is also well-known that if f has non-zero linear structures, then there exists a nonsingular nn matrix B over GF () such that f(xb) =g(y) h(z), where x =(y z), x V n, y V p, z V q, p + q = n, g
4 is a function on V p and g has no non-zero linear structures, h is a linear function on V q. Hence q is equal to the linearity off. There exist a number of equivalent denitions of correlation immune functions [1, 4]. It is easy to verify that the following denition is equivalent to Denition.1 of [1]: Denition 6. Let f be a function on V n and let be its sequence. Then f is called a kth-order correlation immune function if and only if h `i = 0 for every `, the sequence of a linear function '(x) = h xi on V n constrained by 1 W () k. The following lemma is the re-statement of a relation proved in Section of []. Lemma. For every function f on V n,wehave (( 0 ) ( 1 ) ::: ( ))H n =(h `0i h `1i ::: h ` i ): where `i is the ith row of H n, j =0 1 ::: n ; 1. 3 Bent Functions and Partially-bent Functions Notation 1 Let f be a function on V n, the sequence of f and `i denote the ith row of H n, i =0 1 ::: n ; 1. Set = = fi j 0 i n ; 1 h `ii 6= 0g, < = f j () 6= 0 V n g and M =maxfj()jj V n 6= 0g It is easy to verify that #=, #< and M are invariant under any nonsingular linear transformation on the variables, where # denotes the cardinal number of aset. P Since h `ji = n (Parseval's equation, Page 416, [5]) and ( 0 )= n, neither = nor < is an empty set.= reects the correlation immunity property off, while< reects its propagation characteristics and M forecasts the avalanche property of the function. Therefore information on #=, #< and M is useful in determining important cryptographic characteristics of f. Denition 7. A function f on V n is called a bent function [8] if h `ii = n for every i =0 1 ::: n ; 1, where `i is the ith row of H n, i =0 1 ::: n ; 1. Abent function on V n exists only when n is even, and it achieves the maximum nonlinearity n;1 ; 1 n;1.from [8] and Parseval's equation, we have the following: Theorem 1. Let f be a function on V n and denote the sequence off. Thenthe following statements are equivalent: (i) f is bent, (ii) for each i, 0 i n ; 1, h `ii = n where `i is the ith row of H n, i =0 1 :::, (iii) #< =1, (iv) M =0, (v) the nonlinearity of f, N f, satises n;1 ; 1 n;1, (vi) the matrix of f is an Hadamard matrix.
5 An interesting theorem of [] explores a relationship between #= and #<. This result can be expressed as follows. Theorem. For any function f on V n, we have (#=)(#<) n, where the equality holds if and only if there exists a nonsingular nn matrix B over GF () andavector V n such that f(xb ) =g(y)h(z), where x =(y z), x V n, y V p, z V q, p + q = n, g is a bent on V p and h is a linear function on V q. Based on the above theorem, the concept of partially-bent functions was also introduced in the same paper []. Denition 8. A function on V n is called apartially-bent function if (#=)(#<) = n. One can see that partially-bent functions include both bent functions and ane functions. Applying Theorem together with properties of linear structures, or using Theorem of [10] directly, we have Proposition 1. A function f on V n is a partially-bent function if and only if each j()j takes the value of n or 0 only. Equivalently, f is a partially-bent function if and only if < is composed of linear structures. Some partially-bent functions have a high nonlinearity and satisfy the SAC or the propagation criterion of a high degree. Furthermore, some partially-bent functions are balanced. All these properties are useful in cryptography. 4 Plateaued Functions Now we introduce a new class of functions called plateaued functions. Here is the denition. Denition 9. Let f be a function on V n and denote the sequence off. Ifthere exists an even number r, 0 r n, such that #= = r and each h `ji takes the value of n;r or 0 only, where `j denotes the jth row of H n, j =0 1 :::, then f is called arth-order plateaued function on V n. f is also called aplateaued function on V n if we ignore the particular order r. Due to Parseval's equation, the condition #= = r can be obtained from the condition \each h `ji takes the value of n;r or 0 only, where `j denotes the jth row ofh n, j =0 1 ::: n ; 1". For the sake ofconvenience, however, we mentioned both conditions in Denition 9. The following result can be obtained immediately from Denition 9. Proposition. Let f be a function on V n. Then we have (i) if f is a rth-order plateaued function then r must be even, (ii) f is an nth-order plateaued function if and only if f is bent, (iii) f is a 0th-order plateaued function if and only if f is ane.
6 Th following is a consequence of Theorem 3 of [10]. Proposition 3. Every partially-bent function is a plateaued function. In the coming sections we characterize plateaued functions and disprove the converse of Proposition 3. 5 Characterizations of Plateaued Functions First we introduce Holder's Inequality [3]. It states that for real numbers a j 0, b P j 0, j =1 ::: k, P p and q P with p>1and p q = 1, the following is true: k ( j=1 ap j )1=p k ( j=1 bq j )1=q k j=1 a jb j where the quality holds if and only if there exists a constant 0suchthata j = b j for each j =1 ::: k. In particular, set p = q = inholder's Inequality. We conclude kx j=1 a j b j vut ( kx j=1 kx a j )( b j ) (1) where the quality holds if and only if there exists a constant 0 such that a j = b j for each j =1 ::: k. Notation Let f be a function on V n and denote the sequence of f. Let denote the real valued (0 1)-sequence dened as = (c 0 c 1 ::: c ) where c j = 1 if j = 0 otherwise and j V n, that is the binary representation of integer j. Write where each s j is an integer. We note that 6 4 h `0i h `1i. h ` i j=1 H n =(s 0 s 1 ::: s ) () P = h `ji = n where the second equality holds thanks to Parseval's equation. By using Lemma, wehave H n n. Noticing ( 0 )= n,weobtains 0 n + P j=1 s j( j )= n. Since 6 4 ( 0 ) ( 1 ) ( j )=0if j 6< (3) = ( ) P s 0 n + j < j>0 s j ( j P )= n. As s 0 =#=, where # denotes the cardinal number of a set, we have j < j>0 s j ( j )= n ( n X ; #=). Note that n ( n ; #=) = X s j ( j ) js j ( j )js M M (#<;1) (4) j < j>0 j < j>0
7 Hence the following inequality holds. s M M (#<;1) n ( n ; #=) (5) From (), #= n = X s j or #=( n ; #=) = X j=1 s j (6) Theorem 3. Let f be a function on V n and denote the sequence off. Then X ( j ) 3n #= where the equality holds if and only if f is a plateaued function. Proof. By using (4), (1) and (6), we obtain n X j < s j ( j ) X vut X ( n X ;1 s j )( j < js j ( j )j v ut#= X ( j )) n s( X j < s j )( X j < ( j )) ( j ) (7) P Hence 3n #= ( j ). We have proved the inequality P in the theorem. Assume that the equality in the theorem holds i.e., ( j ) = 3n. #= This implies that all the qualities in (7) hold. Hence X X n = s j ( j )= js j ( j )j = j < j < = vut X ( n X ;1 s j )( v ut#= X ( j )) = n s( X j < s j )( X j < ( j )) ( j ) (8) Applying the property of Holder's Inequality to(8),we conclude that j( j )j = js j j j < (9) where >0 is a constant. Applying (9) and (6) to (8), we have n = X j < js j ( j )j = vut X #= n s j = #=n (10)
8 From (10), we have P j < s j ( j )= P j < js j ( j )j. Hence (9) can be expressed more accurately as follows ( j )=s j j < (11) where >0isaconstant. From (8), it is easy to see that P j < s j = P s j. Hence Combining (11), (1) and (3), we have s j =0if j 6< (1) (s 0 s 1 ::: s )=(( 0 ) ( 1 ) ::: ( )) (13) Comparing (13) and (), we obtain H n =(( 0 ) ( 1 ) ::: ( )) (14) Further comparing (14) and the equation in Lemma, we obtain n =(h `0i ::: h ` i ) (15) Noticing that is a real valued (0 1)-sequence, containing #= ones and by using Parseval's equation, we obtain n (#=) = n. Hence (#=) = n, and there exists an integer r with 0 r n such that#= = r and = n;r.from (15) it is easy to see that h `ji = n;r or 0. Hence r must be even. This proves that f is a plateaued function. Conversely assume that f is a plateaued function. Then there exists an even number r, 0 r n, such that #= = r and h `ji = n;r or 0, Due to Lemma, we have P ( j ) = ;n P h `ji 4 = ;n r 4n;r = 3n;r. Hence we have proved P ( j )= 3n #=. Lemma 3. Let f be a function on V n and denote the sequence off. Then the nonlinearity N f of f satises N f n;1 ; p#= n;1,where the equality holds if and only if f is a plateaued function. Proof. Set p M =maxfjh `jijj j =0 1 ::: n ; 1g, where `j is the jth row of H n,0j n ; 1. Using Parseval's equation, we obtain p M #=n. Due to Lemma 1, N f n;1 ; p#= n;1. Assume that f is a plateaued function. Then there exists an even number r, 0 r n, such that #= = r and each h `ji takes either the value of n;r or 0 only, where `j denotes the jth row ofh n, j =0 1 ::: n ; 1. Hence p M = n; 1 r. By using Lemma 1, we have N f = n;1 ; n; 1 r;1 = n;1 ; n;1 p#=. Conversely assume that N f = n;1 ; p#= n;1. From Lemma 1, we have also N f = n;1 ; 1 p p M. Hence p M #= = n. Since both p M and p #= are integers and powers of two, we can let #= = r, where r is an integer with 0 r P n. Hence p M = n; r. Obviously r is even. From Parseval's equation, j= h `ji = n, and the fact that p M #= = n, we conclude that h `ji = n;r for all j =. This proves that f is a plateaued function.
9 From the proof of Lemma 3, we can see that Lemma 3 can be stated in a dierent way as follows. Lemma 4. Let f be a function f on V n and denote the sequence of f. Set p M = maxfjh `jijj j = 0 1 ::: n ; 1g, where `j is the jth row of H n, 0 j n ; 1. Then p M p #= n where the equality holds if and only if f is a plateaued function. Summarizing Theorem 3, Lemmas 3 and 4, we conclude Theorem 4. Let f be a function on V n and denote the sequence of f. Set p M = maxfjh `jijj j = 0 1 ::: n ; 1g, where `j is the jth row of H n, 0 j n ; 1. Then the following statements are equivalent: (i) f is a plateaued function on V n, (ii) there exists an even number r, 0 r n, such that #= = r and each h `ji takes the value of n;r P or 0 only, where `j denotes the jth row of H n, j = 0 1 ::: n ; 1, (iii) ( j ) = 3n, (iv) the #= nonlinearity of f, N f, satises N f = n;1 ; p#= n;1, (v) p p M #= = n, (vi) N f = n;1 ; ; n ;1 q P ( j ). Proof. Due to Denition 9, Theorem 3, Lemmas 3 and 4, (i), (ii), (iii), (iv) and (v) hold. (vi) follows from (iii) and (iv). Theorem 5. Let f be afunction on V n and denote the sequence off. Then the nonlinearity N f of f satises v N f n;1 ; ; n ut n X ;1 ;1 ( j ) where the equality holds if and only if f is a plateaued function on V n. Proof. Set p M = maxfjh `jijj j =0 1 ::: n ; 1g. Multiplying the equality in P h `ji. Lemma by itself, wehave n P ( j )= P h `ji 4 p M Applying Parseval's equation to the above equality, we have P ( j ) n p M. Hence p M ; n q P ( j ). By using Lemma 1, we have proved the inequality N f n;1 ; ; n ;1 q P ( j ). The rest part of the theorem can be proved by using Theorem 4. Theorem 3, Lemmas 3 and 4 and Theorem 4 represent characterizations of plateaued functions. To close this section, let us note that since ( 0 ) = n and #= n, we have n;1 ; ; n ;1 q P ( j ) n;1 ; n ;1 and n;1 ; n;1 p#= n;1 ; n ;1. Hence both inequalities N f n;1 ; ; n ;1 q P ( j )and N f n;1 ; p#= n;1 N f n;1 ; n ;1. are improvements on a more commonly used inequality
10 6 Other Cryptographic Properties of Plateaued Functions By using Lemma 1, we conclude Proposition 4. Let f be arth-order plateaued function on V n. Then the nonlinearity N f of f satises N f = n;1 ; n; r ;1. The following result is the same as Theorem 18 of [13]. Lemma 5. Let f be a function on V n (n ), be the sequence off, andp is an integer, p n. Ifh `ji 0 (mod n;p+ ), where `j is the jth row of H n, j =0 1 ::: n ; 1, then the degree off is at most p ; 1. Using Lemma 5, we obtain Proposition 5. Let f be arth-order plateaued function on V n. Then the algebraic degree off, denoted bydeg(f), satises deg(f) r +1. We note that the upper bound on degree in Proposition 5 is tight for r<n. For the case of r = n, the function, mentioned in Proposition 5, is a bent function on V n. [8] gives a better upper bound on degree of bent function on V n.that bound is n. The following property of plateaued functions can be veried by noting their denition. Proposition 6. Let f be arth-order plateaued function on V n, B be any nonsingular n n matrix over GF () and be anyvector in V n. Then f(xb ) is also a rth-order plateaued function on V n. Theorem 6. Let f be arth-order plateaued function on V n. Then the linearity of f, q, satises q n ; r, wheretheequality holds if and only if f is partiallybent. Proof. There exists a nonsingular nn matrix B over GF () such that f(xb) = g(y) h(z), where x =(y z), y V p, z V q, p + q = n, g is a function on V p and g has no non-zero linear structures, h is a linear function on V q. Hence q is equal to the linearity off. Set f (x) =f(xb). Let, and denote the sequences of f, g and h respectively. Itiseasyto verify =, where denotes the Kronecker product [1]. From the structure of H n, each rowofh n, L, can be expressed as L = ` e, where ` is a row ofh p and e is a row ofh q.itiseasytoverify h Li = h `ih ei (16) Since h is linear, is a row ofh q. Replace e by in (16), we have where L 0 = ` is still a row ofh n. h L 0 i = h `ih i = q h `i (17)
11 Note that f is also a rth-order plateaued function on V n. Hence h Li takes the value of n; 1 r or zero only. Due to (17), h `i takes the value of n; 1 r;q = p; 1 r or zero only. This proves that g is a rth-order plateaued function on V p. Hence r p and r n ; q, i.e.,qn ; r. Assume that q = n ; r. Thenp = r. From (17), each h `i takes the value of r = p or zero only, where ` is any rowofh p. Hence applying Parseval's equation to g, we can conclude that for each row ` of H p, h `i cannot take the value of zero. In other words, for each row ` of H p, h `i takes the value of p only. Hence we have proved that g is a bent function on V p. Due to Theorem, f is partially-bent. Conversely, assume that f is partially-bent. Due to Theorem, g is a bent function on V p. Hence each h `i takes the value of p only, where ` is any row ofh p. Note that both and e are rows of H q hence h ei takes the value q or zero only. From (16), we conclude that h Li takes the value q+ p or zero only. Recall f is a rth-order plateaued function on V n. Hence q + p = n ; r. This implies that r = p, i.e., q = n ; r. 7 Relationships between Partially-bent Functions and Plateaued Functions To examine more profound relationships between partially-bent functions and plateaued functions, we introduce one more characterization of partially-bent functions as follows. Theorem 7. For every function f on V n,wehave n ; #= #= M n (#<;1) where the equality holds if and only if f is partially-bent. Proof. From Notation, we have s M s 0 = #=. As a consequence of (5), we obtain the inequality in the theorem. Next we consider the equality in the theorem. Assume that the equality holds, i.e., M (#<;1)#= = n ( n ; #=) (18) From (4), n ( n ; #=) X X j < j>0 js j ( j )j M js j j M (#<;1)#= (19) j < j>0 From (18), we can see that all the equalities in (19) hold. Hence M (#<;1)#= = X j < j>0 js j ( j )j (0)
12 Note that js j j#= and j( j )j M, for j>0. Hence from (0), we obtain js j j =#= whenever j <and j>0 (1) and j( j )j = M for all j <with j>0. P Applying (1) to (6), and noticing s 0 =#=, we obtain #= n = P s j j < s j = (#<)(#=). This results in n (#<)(#=). Together with the inequality intheorem, it proves that (#<)(#=) = n, i.e., f is a partiallybent function. Conversely assume that f is a partially-bent function, i.e., (#=)(#<) = n. Then the inequality in the theorem is specialized as M ( n ; #=) n ( n ; #=) () We need to examine two cases. Case 1: #= = n.obviously the equality in () holds. Case : #= 6= n. From (), we have M n. Thus M = n. This completes the proof. Next we consider a non-bent function f. With such a function wehave M 6= 0. Thus from Theorem 7, we have the following result. Corollary 1. For every non-bent function f on V n, we have (#=)(#<) n ( n ; #=) M +#= where the equality holds if and only if f is partially-bent (but not bent). Proposition 7. For every non-bent function f, we have n ( n ; #=) +#= n M where the equality holds if and only if #= = n or f has a non-zero linear structure. Proof. Since M n, the inequality isobvious. On the other hand, it is easy to see that the equality holds if and only if ( n ; M )( n ; #=) =0. From Proposition 7, one observes that for any non-bent function f, Corollary 1 implies Theorem. Theorem 8. Let f be a rth-order plateaued function. Then the following statements are equivalent: (i) f is a partially-bent function, (ii) #< = n;r, (iii) M (#<;1) = n;r ; n, (iv) the linearity q of f satises q = n ; r. Proof. (i) =) (ii). Since f is a partially-bent function, we have(#=)(#<) = n. As f is a rth-order plateaued function, #= = r and hence #< = n;r.
13 (ii) =) (iii). It is obviously true when r = n. Now consider the case of r<n. Using Theorem 7, we have n ;#= M #= n (#<;1) which is specialized as n;r ; 1 M n (n;r ; 1) (3) From (3) and the fact that M n,we obtain n;r ; 1 M n ( n;r ; 1) n;r ; 1. Hence M = n or r = n. (iii) obviously holds when M = n.when r = n, wehave #< = 1 and hence (iii) also holds. (iii) =) (i). Note that (iii) implies n ;#= = M #= n (#< ;1) where #= = r. By Theorem 7, f is partially-bent. Due to Theorem 6, (iv) () (i). 8 Construction of Plateaued Functions and Disproof of The Converse of Proposition 3 Lemma 6. For any positive integers t and k with k < t < k, there exist t non-zero vectors in V k,say 0 1, :::, t ;1, such that for any non-zero vector V k, the t -set f' 0 () ' 1 () ::: ' ()g, contains both zero andone, t;1 where ' is the linear function on V k dened by' (x) =h xi. Proof. We choose k linearly independent vectors in V k, say 1 ::: k. From linear algebra, (h 1 i ::: h k i) goes through all the non-zero vectors in V k exactly once while goes through all the non-zero vectors in V k. Hence there exists a unique satisfying (h 1 i ::: h k i)=(1 ::: 1). Furthermore, for any non-zero vector V k with 6=, fh 1 i ::: h k ig contains both one and zero. Let 0 be a non-zero vector in V k, such that h 0 i =0.Obviously 0 6 f 1 ::: k g. If t > k +1, choose other t ; k ; 1 non-zero vectors in V k, k+1 ::: t ;1, such that 0 1 ::: k k+1 ::: t ;1 are mutually distinct. It is easy to see that for any non-zero vector V k, f 0 1 ::: t ;1 g contains both one and zero. This proves the lemma. The following example proves the existence of rth-order plateaued functions on V n, where 0 < r < n, and disproves the converse of Proposition 3. We note that in this section, we will not discuss nth-order and 0th-order plateaued function on V n as they are bent and ane functions respectively. Example 1. Let t and k be positiveintegers with k< t < k.let 0 1 ::: t ;1 be the t non-zero vectors in V k dened in Lemma 6. Let j denote the sequence of ' j, j =0 1 ::: t ; 1. Set = 0 1 ::: t ;1.Letn = k + t and f be the function on V n whose sequence is. By using the properties of H n, it is easy to verify that each h `ji takes the value of k or 0 only, where `i is the ith row ofh n, i =0 1 ::: n ; 1. Using Parseval's equation, we obtain#= = n;k.letr =n ; k =t. Then f is a rth-order plateaued function on V n.dueton = k + t, r =n ; k =t and t<k,0<r<nholds.
14 We now consider () with the function f. Let = ( ) where V t, V k.notethat () = (P ji= h j i ()i if 6= 0 P t ;1 h j j ()i if = 0 but 6= 0 (4) where j V t is the binary representation of an integer j, j =0 1 ::: n ; 1. Since ' j 6= ' i if j 6= i, where ' (x) =h xi, ' j (x) ' i (x ) is a nonzero linear function and hence balanced. We have nowproved h j i ()i = 0 for j 6= i. Hence () = 0 when 6= 0. On the other hand, for any linear function ' on V k,wehave '(x) '(x ) = '(). Hence h j j ()i = k if and only if ' j () = 0. In addition, h j j ()i = ; k if and only if ' j () = 1. By using Lemma 6, we have () = P t ;1 h j j ()i 6= t k = n for 6= 0. In summary, we have () 8 < : =0 if 6= 0 6= n if = 0 and 6= 0 = n if =0 (5) Since f is a rth-order plateaued function on V n and r<n, f is not bent, on the other hand, (5) shows that f has non-zero linear structures. Hence we conclude that f is not partially-bent. Hence we have proved that f is plateaued but not partially-bent. This disproves the converse of Proposition 3. f has some other interesting properties. In particular, due to Proposition 4, the nonlinearity N f of f satises N f = n;1 ; n; r ;1. Note that the sequenceofany non-zero linear function is (1 ;1)-balanced. Hence each j and = 0 1 ::: t ;1 are (1 ;1)-balanced. This implies that f is (0 1)-balanced. Since the function f is not partially-bent, by using Theorem, we have (#=)(#<) > n. This proves that #< > n;r. On the other hand, from (5), we have #< k = n; 1 r.thus we can conclude that n;r < #< n; 1 r. We end this example by noting that such functions as f exist on V n both for n even and odd. Now we summarize the relationships among bent, partially-bent and plateaued functions. Let B n denote the set of bent functions on V n, P n denote the set of partially-bent functions on V n and F n denote the set of plateaued functions on V n. Then the above results imply that B n P n F n, where denotes the relationship of proper subset. We further let G n denote the set of plateaued functions on V n that do not have non-zero linear structures and are not bent functions. The relationships among these classes of functions are shown in Figure 1. Example 1 proves that G n is nonempty. Next we consider how to improve the function in Example 1 so as to obtain a rth-order plateaued function on V n satisfying the SAC and all the properties mentioned in Example 1.
15 Fig. 1. Relationship among bent, partially bent, and plateaued functions Example. Note that if r>, i.e., t>1, then from Example 1, #< n; 1 r < n;1. In other words, #< c > n;1 where < c denotes the complementary set of <. Hence there exist n linearly independent vectors in < c. In other words, there exist n linearly independent vectors with respect to which f satises the propagation criterion. Hence we can choose a nonsingular n n matrix A over GF () such that g(x) =f(xa) satises the SAC (see [9]). The nonsingular linear transformation A does not alter any of the properties of f in Example 1 We can further improve the function in Example so as to obtain a rthorder plateaued functions on V n having the highest degree and satisfying all the properties in Example 1. Example 3. Given any vector =(i 1 ::: i s ) V t,we dene a function on V t by D (y) =(y 1 i1 ) (y t is ) where y =(y 1 ::: y t )and i =1 i indicates the binary complement of i. Let i1i p,(i 1 ::: i p ) V p, be the sequence of a function f i1i p (x 1 ::: x q ) on V q. Let be the concatenation of 000, 001, :::, 111, namely, = ( ::: 111 ). It is easy to verify that is the sequence of a function on V q+p given by f(y 1 ::: y p z 1 ::: z q )= M (i 1i p)vp D i1i p (y 1 ::: y p )f i1i p (z 1 ::: z q ): (6) Let j V t is the binary representation of integer j, j =0 1 ::: t ;1. Write 0 = ' 0, 1 = ' 1, :::, t;1 = ' t;1, where 0 1 ::: t ;1 are the same with those in Example 1 and ' = h xi.
16 Due to (6), the function f on V t+k, mentioned in Example 1 can be expressed as f(y z) = L V t D (y) (z). Case 1: L V t 6= 0. Write L V t =, where must be a non-zero linear function on V k. Note that each D (y) contains y 1 y t. Hence the term y 1 y t (z) survives in the nal algebraic normal form representation of f(y z) and hence the degree of f is t +1= r +1. Case : L V t = 0, i.e., L t ;1 ' j = 0. Note that there exist k ; 1 nonzero vectors in V k and k ; 1 > t. Hence we can replace ' byany non-zero t;1 linear function ' on V k, that diers from ' 0, ' 1, ::: '. This reduces Case t;1 to Case 1. We have now constructed a rth-order plateaued function with degree r + 1. Applying the discussions in Examples 1 and, we can obtain a rth-order plateaued function on V n having degree r + 1 and satisfying all the properties of the function constructed in Example 1. It should be noted that the function in this example achieves the highest possible algebraic degree given in Proposition 4. Thus the upper bound on the algebraic degree of plateaued functions, mentioned in Proposition 5, is tight. 9 Conclusions We have introduced and characterized a new class of functions called plateaued functions. These functions bring together various nonlinear characteristics. We have also shown that partially-bent functions are a proper subset of plateaued functions, which settles an open problem related to partially-bent functions. We have further demonstrated methods for constructing plateaued functions that have many cryptographically desirable properties. Acknowledgement The second author was supported by a Queen Elizabeth II Fellowship ( ). References 1. P. Camion, C. Carlet, P. Charpin, and N. Sendrier. On correlation-immune functions. In Advances in Cryptology - CRYPTO'91, volume 576, Lecture Notes in Computer Science, pages 87{100. Springer-Verlag, Berlin, Heidelberg, New York, Claude Carlet. Partially-bent functions. Designs, Codes and Cryptography, 3:135{ 145, Friedhelm Erwe. Dierential And Integral Calculus. Oliver And Boyd Ltd, Edinburgh And London, Xiao Guo-Zhen and J. L. Massey. A spectral characterization of correlationimmune combining functions. IEEE Transactions on Information Theory, 34(3):569{571, 1988.
17 5. F. J. MacWilliams and N. J. A. Sloane. The Theory of Error-Correcting Codes. North-Holland, Amsterdam, New York, Oxford, W. Meier and O. Staelbach. Nonlinearity criteria for cryptographic functions. In Advances in Cryptology -EUROCRYPT'89, volume 434, Lecture Notes in Computer Science, pages 549{56. Springer-Verlag, Berlin, Heidelberg, New York, B. Preneel, W. V. Leekwijck, L. V. Linden, R. Govaerts, and J. Vandewalle. Propagation characteristics of boolean functions. In Advances in Cryptology - EU- ROCRYPT'90, volume 437, Lecture Notes in Computer Science, pages 155{165. Springer-Verlag, Berlin, Heidelberg, New York, O. S. Rothaus. On \bent" functions. Journal of Combinatorial Theory, Ser. A, 0:300{305, J. Seberry, X. M. Zhang, and Y. Zheng. Improving the strict avalanche characteristics of cryptographic functions. Information Processing Letters, 50:37{41, J. Wang. The linear kernel of boolean functions and partially-bent functions. System Science and Mathematical Science, 10:6{11, A. F. Webster and S. E. Tavares. On the design of S-boxes. In Advances in Cryptology - CRYPTO'85, volume 19, Lecture Notes in Computer Science, pages 53{534. Springer-Verlag, Berlin, Heidelberg, New York, R. Yarlagadda and J. E. Hershey. Analysis and synthesis of bent sequences. IEE Proceedings (Part E), 136:11{13, X. M. Zhang, Y. Zheng, and Hideki Imai. Duality ofboolean functions and its cryptographic signicance. In Advances in Cryptology - ICICS'97, volume 1334, Lecture Notes in Computer Science, pages 159{169. Springer-Verlag, Berlin, Heidelberg, New York, 1997.
Non-Separable Cryptographic Functions
International Symposium on Information Theory and Its Applications Honolulu, Hawaii, USA, November 5 8, 2000 Non-Separable Cryptographic Functions Yuliang Zheng and Xian-Mo Zhang School of Network Computing
More informationStructures of cryptographic functions with strong avalanche characteristics
University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 1995 Structures of cryptographic functions with strong avalanche characteristics
More informationWell known bent functions satisfy both SAC and PC(l) for all l n, b not necessarily SAC(k) nor PC(l) of order k for k 1. On the other hand, balancedne
Design of SAC/PC(l) of order k Boolean functions and three other cryptographic criteria Kaoru Kurosawa 1 and Takashi Satoh?2 1 Dept. of Comper Science, Graduate School of Information Science and Engineering,
More informationComments on "Generating and Counting Binary Bent Sequences"
University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 1994 Comments on "Generating and Counting Binary Bent Sequences" Claude
More informationOn the Symmetric Property of Homogeneous Boolean Functions
On the Symmetric Property of Homogeneous Boolean Functions Chengxin Qu, Jennifer Seberry, and Josef Pieprzyk Centre for Computer Security Research School of Information Technology and Computer Science
More informationHadamard Matrices, Bent Functions and Cryptography. The University of Wollongong. Abstract
Hadamard Matrices, Bent Functions and Cryptography Jennifer Seberry and ian-mo Zhang Department of Computer Science The University of Wollongong Wollongong, NSW 5, AUSTRALIA E-mail: fj.seberry,xianmog@cs.uow.edu.au
More informationNonlinearity and propagation characteristics of balanced boolean functions
University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 1994 Nonlinearity and propagation characteristics of balanced boolean
More informationarxiv: v5 [cs.it] 4 Nov 2009
1 Constructions of Almost Optimal Resilient Boolean Functions on Large Even Number of Variables WeiGuo ZHANG and GuoZhen XIAO ISN Lab, Xidian University, Xi an 710071, P.R.China arxiv:0905.0794v5 [cs.it]
More informationOn Cryptographic Properties of the Cosets of R(1;m)
1494 IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 47, NO. 4, MAY 2001 On Cryptographic Properties of the Cosets of R(1;m) Anne Canteaut, Claude Carlet, Pascale Charpin, and Caroline Fontaine Abstract
More informationDecomposing Bent Functions
2004 IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 49, NO. 8, AUGUST 2003 Decomposing Bent Functions Anne Canteaut and Pascale Charpin Abstract In a recent paper [1], it is shown that the restrictions
More informationarxiv:math/ v1 [math.co] 24 Oct 2000
arxiv:math/0010220v1 [math.co] 24 Oct 2000 Nonlinearity, Local and Global Avalanche Characteristics of Balanced Boolean Functions Abstract Pantelimon Stănică Auburn University Montgomery, Department of
More informationSome results concerning global avalanche characteristics of two q-ary functions
Some results concerning global avalanche characteristics of two -ary functions Brajesh Kumar Singh Department of Mathematics, School of Allied Sciences, Graphic Era Hill University, Dehradun-4800 (Uttarakhand)
More informationPitfalls in designing substitution boxes
University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 1994 Pitfalls in designing substitution boxes Jennifer Seberry University
More informationMaximum Correlation Analysis of Nonlinear S-boxes in Stream Ciphers
Maximum Correlation Analysis of Nonlinear S-boxes in Stream Ciphers Muxiang Zhang 1 and Agnes Chan 2 1 GTE Laboratories Inc., 40 Sylvan Road LA0MS59, Waltham, MA 02451 mzhang@gte.com 2 College of Computer
More informationThird-order nonlinearities of some biquadratic monomial Boolean functions
Noname manuscript No. (will be inserted by the editor) Third-order nonlinearities of some biquadratic monomial Boolean functions Brajesh Kumar Singh Received: April 01 / Accepted: date Abstract In this
More informationSmart Hill Climbing Finds Better Boolean Functions
Smart Hill Climbing Finds Better Boolean Functions William Millan, Andrew Clark and Ed Dawson Information Security Research Centre Queensland University of Technology GPO Box 2434, Brisbane, Queensland,
More information1-Resilient Boolean Function with Optimal Algebraic Immunity
1-Resilient Boolean Function with Optimal Algebraic Immunity Qingfang Jin Zhuojun Liu Baofeng Wu Key Laboratory of Mathematics Mechanization Institute of Systems Science, AMSS Beijing 100190, China qfjin@amss.ac.cn
More informationnonlinearities to resist certain attacks on these ciphers (correlation and linear attacks). A Boolean function is called bent if its nonlinearity equa
Upper bounds on the numbers of resilient functions and of bent functions Claude Carlet 1 and Andrew Klapper 2 1 INRIA projet CODES, B.P. 105, 78153 Le Chesnay Cedex- France. Claude.Carlet@inria.fr 2 Dept.
More informationConstruction of 1-Resilient Boolean Functions with Optimal Algebraic Immunity and Good Nonlinearity
Pan SS, Fu XT, Zhang WG. Construction of 1-resilient Boolean functions with optimal algebraic immunity and good nonlinearity. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 26(2): 269 275 Mar. 2011. DOI 10.1007/s11390-011-1129-4
More informationHadamard Matrices, d-linearly Independent Sets and Correlation-Immune Boolean Functions with Minimum Hamming Weights
Hadamard Matrices, d-linearly Independent Sets and Correlation-Immune Boolean Functions with Minimum Hamming Weights Qichun Wang Abstract It is known that correlation-immune (CI) Boolean functions used
More informationAlgebraic nonlinearity and its applications to cryptography
Algebraic nonlinearity and its applications to cryptography Luke O Connor Department of Computer Science University of Waterloo, Ontario, Canada, NL 3G1 Andrew Klapper Department of Computer Science University
More informationTransform Domain Analysis of DES. Guang Gong and Solomon W. Golomb. University of Southern California. Tels and
Transform Domain Analysis of DES Guang Gong and Solomon W. Golomb Communication Sciences Institute University of Southern California Electrical Engineering-Systems, EEB # 500 Los Angeles, California 90089-2565
More informationGeneralized hyper-bent functions over GF(p)
Discrete Applied Mathematics 55 2007) 066 070 Note Generalized hyper-bent functions over GFp) A.M. Youssef Concordia Institute for Information Systems Engineering, Concordia University, Montreal, QC, H3G
More informationA New Class of Bent Negabent Boolean Functions
A New Class of Bent Negabent Boolean Functions Sugata Gangopadhyay and Ankita Chaturvedi Department of Mathematics, Indian Institute of Technology Roorkee Roorkee 247667 INDIA, {gsugata, ankitac17}@gmail.com
More informationSome properties of q-ary functions based on spectral analysis
Some properties of q-ary functions based on spectral analysis Deep Singh and Maheshanand Bhaintwal Department of Mathematics, Indian Institute of Technology Roorkee, Roorkee 247667 INDIA deepsinghspn@gmail.com,mahesfma@iitr.ernet.in
More informationA Conjecture on Binary String and Its Applications on Constructing Boolean Functions of Optimal Algebraic Immunity
A Conjecture on Binary String and Its Applications on Constructing Boolean Functions of Optimal Algebraic Immunity Ziran Tu and Yingpu deng Abstract In this paper, we propose a combinatoric conjecture
More informationCryptographically Robust Large Boolean Functions. Debdeep Mukhopadhyay CSE, IIT Kharagpur
Cryptographically Robust Large Boolean Functions Debdeep Mukhopadhyay CSE, IIT Kharagpur Outline of the Talk Importance of Boolean functions in Cryptography Important Cryptographic properties Proposed
More informationSome results on the existence of t-all-or-nothing transforms over arbitrary alphabets
Some results on the existence of t-all-or-nothing transforms over arbitrary alphabets Navid Nasr Esfahani, Ian Goldberg and Douglas R. Stinson David R. Cheriton School of Computer Science University of
More informationSecret-sharing with a class of ternary codes
Theoretical Computer Science 246 (2000) 285 298 www.elsevier.com/locate/tcs Note Secret-sharing with a class of ternary codes Cunsheng Ding a, David R Kohel b, San Ling c; a Department of Computer Science,
More informationNew Constructions for Resilient and Highly Nonlinear Boolean Functions
New Constructions for Resilient and Highly Nonlinear Boolean Functions Khoongming Khoo 1 and Guang Gong 2 1 Department of Combinatorics and Optimization, 2 Department of Electrical and Computer Engineering,
More informationOn Existence and Invariant of Algebraic Attacks
On Existence and Invariant of Algebraic Attacks Guang Gong Department of Electrical and Computer Engineering University of Waterloo Waterloo, Ontario N2L 3G1, CANADA Email. ggong@calliope.uwaterloo.ca
More informationElementary 2-Group Character Codes. Abstract. In this correspondence we describe a class of codes over GF (q),
Elementary 2-Group Character Codes Cunsheng Ding 1, David Kohel 2, and San Ling Abstract In this correspondence we describe a class of codes over GF (q), where q is a power of an odd prime. These codes
More informationConstruction and Count of Boolean Functions of an Odd Number of Variables with Maximum Algebraic Immunity
arxiv:cs/0605139v1 [cs.cr] 30 May 2006 Construction and Count of Boolean Functions of an Odd Number of Variables with Maximum Algebraic Immunity Na Li, Wen-Feng Qi Department of Applied Mathematics, Zhengzhou
More informationHighly nonlinear 0-1 balanced boolean functions satisfying strict avalanche criterion
University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 1993 Highly nonlinear 0-1 balanced boolean functions satisfying strict
More informationBoolean Inner-Product Spaces and Boolean Matrices
Boolean Inner-Product Spaces and Boolean Matrices Stan Gudder Department of Mathematics, University of Denver, Denver CO 80208 Frédéric Latrémolière Department of Mathematics, University of Denver, Denver
More informationFast Algebraic Immunity of 2 m + 2 & 2 m + 3 variables Majority Function
Fast Algebraic Immunity of 2 m + 2 & 2 m + 3 variables Majority Function Yindong Chen a,, Fei Guo a, Liu Zhang a a College of Engineering, Shantou University, Shantou 515063, China Abstract Boolean functions
More informationThe Analysis of affinely Equivalent Boolean Functions
The Analysis of affinely Equivalent Boolean Functions Qing-shu Meng Min Yang Huan-guo Zhang Yuzhen Liu October 21, 2005 Abstract By Walsh transform, autocorrelation function, decomposition, derivation
More informationFinding Low Degree Annihilators for a Boolean Function Using Polynomial Algorithms
Finding Low Degree Annihilators for a Boolean Function Using Polynomial Algorithms Vladimir Bayev Abstract. Low degree annihilators for Boolean functions are of great interest in cryptology because of
More informationHeuristic Design of Cryptographically Strong Balanced Boolean Functions
Heuristic Design of Cryptographically Strong Balanced Boolean Functions William Millan, Andrew Clark and Ed Dawson Information Security Research Center, Queensland University of Technology, GPO Box 2434,
More informationDirichlet Product for Boolean Functions
Dirichlet Product for Boolean Functions Abderrahmane Nitaj 1, Willy Susilo 2 and Joseph Tonien 2 1 Laboratoire de Mathématiques Nicolas Oresme, Université de Caen Normandie, France 2 Centre for Computer
More informationBalanced Boolean Functions with (Almost) Optimal Algebraic Immunity and Very High Nonlinearity
Balanced Boolean Functions with (Almost) Optimal Algebraic Immunity and Very High Nonlinearity Xiaohu Tang 1, Deng Tang 1, Xiangyong Zeng and Lei Hu 3 In this paper, we present a class of k-variable balanced
More informationA New Characterization of Semi-bent and Bent Functions on Finite Fields
A New Characterization of Semi-bent and Bent Functions on Finite Fields Khoongming Khoo DSO National Laboratories 20 Science Park Dr S118230, Singapore email: kkhoongm@dso.org.sg Guang Gong Department
More informationAlgebra II. Paulius Drungilas and Jonas Jankauskas
Algebra II Paulius Drungilas and Jonas Jankauskas Contents 1. Quadratic forms 3 What is quadratic form? 3 Change of variables. 3 Equivalence of quadratic forms. 4 Canonical form. 4 Normal form. 7 Positive
More informationQuadratic Almost Perfect Nonlinear Functions With Many Terms
Quadratic Almost Perfect Nonlinear Functions With Many Terms Carl Bracken 1 Eimear Byrne 2 Nadya Markin 3 Gary McGuire 2 School of Mathematical Sciences University College Dublin Ireland Abstract We introduce
More informationPerfect Algebraic Immune Functions
Perfect Algebraic Immune Functions Meicheng Liu, Yin Zhang, and Dongdai Lin SKLOIS, Institute of Information Engineering, CAS, Beijing 100195, P. R. China meicheng.liu@gmail.com, zhangy@is.iscas.ac.cn,
More informationCorrecting Codes in Cryptography
EWSCS 06 Palmse, Estonia 5-10 March 2006 Lecture 2: Orthogonal Arrays and Error- Correcting Codes in Cryptography James L. Massey Prof.-em. ETH Zürich, Adjunct Prof., Lund Univ., Sweden, and Tech. Univ.
More informationBoolean functions in cryptography
University of Wollongong Research Online University of Wollongong Thesis Collection University of Wollongong Thesis Collections Boolean functions in cryptography Cheng-Xin Qu University of Wollongong Recommended
More informationAtypical usage of one-way hash functions for data integrityisasfollows. The hash-value corresponding to a particular message M is computed at time t 1
A Fast Cryptographic Hash Function Based on Linear Cellular Automata over GF(q) Miodrag Mihaljevic 1, Yuliang Zheng 2 and Hideki Imai 3 1 Mathematical Institute, Serb. Acad. Sci. & Arts Kneza Mihaila 35,
More informationOn Welch-Gong Transformation Sequence Generators
On Welch-Gong Transformation Sequence Generators G. Gong and A.M. Youssef Center for Applied Cryptographic Research, Department of Combinatorics and Optimization, University of Waterloo, Waterloo, Ontario
More informationStatistical and Linear Independence of Binary Random Variables
Statistical and Linear Independence of Binary Random Variables Kaisa Nyberg Department of Computer Science, Aalto University School of Science, Finland kaisa.nyberg@aalto.fi October 10, 2017 Abstract.
More informationGeneralized Correlation Analysis of Vectorial Boolean Functions
Generalized Correlation Analysis of Vectorial Boolean Functions Claude Carlet 1, Khoongming Khoo 2, Chu-Wee Lim 2, and Chuan-Wen Loe 2 1 University of Paris 8 (MAATICAH) also with INRIA, Projet CODES,
More informationImproved Zero-sum Distinguisher for Full Round Keccak-f Permutation
Improved Zero-sum Distinguisher for Full Round Keccak-f Permutation Ming Duan 12 and Xuejia Lai 1 1 Department of Computer Science and Engineering, Shanghai Jiao Tong University, China. 2 Basic Courses
More informationConstructions of Resilient S-Boxes with Strictly Almost Optimal Nonlinearity Through Disjoint Linear Codes
IEEE TRANSACTIONS ON INFORMATION THEORY, VOL 60, NO 3, 2014 1 Constructions of Resilient S-Boxes with Strictly Almost Optimal Nonlinearity Through Disjoint Linear Codes Wei-Guo Zhang, Member, IEEE, and
More informationAn average case analysis of a dierential attack. on a class of SP-networks. Distributed Systems Technology Centre, and
An average case analysis of a dierential attack on a class of SP-networks Luke O'Connor Distributed Systems Technology Centre, and Information Security Research Center, QUT Brisbane, Australia Abstract
More informationUniversity of Bergen Faculty of Mathematical and Natural Sciences Department of Informatics The Selmer Center
University of Bergen Faculty of Mathematical and Natural Sciences Department of Informatics The Selmer Center A DATABASE FOR BOOLEAN FUNCTIONS AND CONSTRUCTIONS OF GENERALIZED COMPLEMENTARY PAIRS by Mohamed
More informationHaar Spectrum of Bent Boolean Functions
Malaysian Journal of Mathematical Sciences 1(S) February: 9 21 (216) Special Issue: The 3 rd International Conference on Mathematical Applications in Engineering 21 (ICMAE 1) MALAYSIAN JOURNAL OF MATHEMATICAL
More information9-variable Boolean Functions with Nonlinearity 242 in the Generalized Rotation Class
9-variable Boolean Functions with Nonlinearity 242 in the Generalized Rotation Class Selçuk Kavut and Melek Diker Yücel arxiv:0808.0684v1 [cs.cr] 5 Aug 2008 Abstract In 2006, 9-variable Boolean functions
More informationConstructions of Resilient S-Boxes with Strictly Almost Optimal Nonlinearity Through Disjoint Linear Codes
IEEE TRANSACTIONS ON INFORMATION THEORY, VOL 60, NO 3, PP 1638-1651, 2014 1 Constructions of Resilient S-Boxes with Strictly Almost Optimal Nonlinearity Through Disjoint Linear Codes Wei-Guo Zhang, Member,
More informationVisual cryptography schemes with optimal pixel expansion
Theoretical Computer Science 369 (2006) 69 82 wwwelseviercom/locate/tcs Visual cryptography schemes with optimal pixel expansion Carlo Blundo a,, Stelvio Cimato b, Alfredo De Santis a a Dipartimento di
More information16 Chapter 3. Separation Properties, Principal Pivot Transforms, Classes... for all j 2 J is said to be a subcomplementary vector of variables for (3.
Chapter 3 SEPARATION PROPERTIES, PRINCIPAL PIVOT TRANSFORMS, CLASSES OF MATRICES In this chapter we present the basic mathematical results on the LCP. Many of these results are used in later chapters to
More informationVisual Cryptography Schemes with Optimal Pixel Expansion
Visual Cryptography Schemes with Optimal Pixel Expansion Carlo Blundo, Stelvio Cimato and Alfredo De Santis Dipartimento di Informatica ed Applicazioni Università degli Studi di Salerno, 808, Baronissi
More informationOn the Arithmetic Walsh Coefficients of Boolean Functions
Designs, Codes, and Cryptography manuscript No. (will be inserted by the editor) On the Arithmetic Walsh Coefficients of Boolean Functions Claude Carlet Andrew Klapper Received: date / Accepted: date Abstract
More informationBalanced Boolean Function on 13-variables having Nonlinearity strictly greater than the Bent Concatenation Bound
Balanced Boolean Function on 13-variables having Nonlinearity strictly greater than the Bent Concatenation Bound Subhamoy Maitra Applied Statistics Unit, Indian Statistical Institute, 203 B T Road, Kolkata
More informationOn Construction of a Class of. Orthogonal Arrays
On Construction of a Class of Orthogonal Arrays arxiv:1210.6923v1 [cs.dm] 25 Oct 2012 by Ankit Pat under the esteemed guidance of Professor Somesh Kumar A Dissertation Submitted for the Partial Fulfillment
More informationLacunary Polynomials over Finite Fields Course notes
Lacunary Polynomials over Finite Fields Course notes Javier Herranz Abstract This is a summary of the course Lacunary Polynomials over Finite Fields, given by Simeon Ball, from the University of London,
More informationSpurious Chaotic Solutions of Dierential. Equations. Sigitas Keras. September Department of Applied Mathematics and Theoretical Physics
UNIVERSITY OF CAMBRIDGE Numerical Analysis Reports Spurious Chaotic Solutions of Dierential Equations Sigitas Keras DAMTP 994/NA6 September 994 Department of Applied Mathematics and Theoretical Physics
More informationFourier Spectra of Binomial APN Functions
Fourier Spectra of Binomial APN Functions arxiv:0803.3781v1 [cs.dm] 26 Mar 2008 Carl Bracken Eimear Byrne Nadya Markin Gary McGuire March 26, 2008 Abstract In this paper we compute the Fourier spectra
More informationDetailed Proof of The PerronFrobenius Theorem
Detailed Proof of The PerronFrobenius Theorem Arseny M Shur Ural Federal University October 30, 2016 1 Introduction This famous theorem has numerous applications, but to apply it you should understand
More informationCCZ-equivalence and Boolean functions
CCZ-equivalence and Boolean functions Lilya Budaghyan and Claude Carlet Abstract We study further CCZ-equivalence of (n, m)-functions. We prove that for Boolean functions (that is, for m = 1), CCZ-equivalence
More informationRANK AND PERIMETER PRESERVER OF RANK-1 MATRICES OVER MAX ALGEBRA
Discussiones Mathematicae General Algebra and Applications 23 (2003 ) 125 137 RANK AND PERIMETER PRESERVER OF RANK-1 MATRICES OVER MAX ALGEBRA Seok-Zun Song and Kyung-Tae Kang Department of Mathematics,
More informationChapter 3 Transformations
Chapter 3 Transformations An Introduction to Optimization Spring, 2014 Wei-Ta Chu 1 Linear Transformations A function is called a linear transformation if 1. for every and 2. for every If we fix the bases
More informationCirculant Hadamard matrices as HFP-codes of type C 4n C 2. arxiv: v1 [math.co] 26 Nov 2017
Circulant Hadamard matrices as HFP-codes of type C 4n C 2. arxiv:1711.09373v1 [math.co] 26 Nov 2017 J. Rifà Department of Information and Communications Engineering, Universitat Autònoma de Barcelona October
More informationSequences, DFT and Resistance against Fast Algebraic Attacks
Sequences, DFT and Resistance against Fast Algebraic Attacks Guang Gong Department of Electrical and Computer Engineering University of Waterloo Waterloo, Ontario N2L 3G1, CANADA Email. ggong@calliope.uwaterloo.ca
More informationVector Space Basics. 1 Abstract Vector Spaces. 1. (commutativity of vector addition) u + v = v + u. 2. (associativity of vector addition)
Vector Space Basics (Remark: these notes are highly formal and may be a useful reference to some students however I am also posting Ray Heitmann's notes to Canvas for students interested in a direct computational
More informationRank and Kernel of binary Hadamard codes.
1 Rank and Kernel of binary Hadamard codes. K.T. Phelps, J. Rifà Senior Member IEEE, M. Villanueva Abstract In this paper the rank and the dimension of the kernel for (binary) Hadamard codes of length
More informationProvable Security Against Differential and Linear Cryptanalysis
Provable Security Against Differential and Linear Cryptanalysis Kaisa Nyberg Aalto University School of Science and Nokia, Finland kaisa.nyberg@aalto.fi Abstract. In this invited talk, a brief survey on
More informationOn Boolean Functions with Generalized Cryptographic Properties
On Boolean Functions with Generalized Cryptographic Properties An Braeken 1, Ventzislav Nikov 2, Svetla Nikova 1, and Bart Preneel 1 1 Department Electrical Engineering, ESAT/COSIC, Katholieke Universiteit
More informationOn Locating-Dominating Codes in Binary Hamming Spaces
Discrete Mathematics and Theoretical Computer Science 6, 2004, 265 282 On Locating-Dominating Codes in Binary Hamming Spaces Iiro Honkala and Tero Laihonen and Sanna Ranto Department of Mathematics and
More informationSubstitution-Permutation Networks Resistant to Differential and Linear Cryptanalysis
J. Cryptology (1996) 9: 1 19 1996 International Association for Cryptologic Research Substitution-Permutation Networks Resistant to Differential and Linear Cryptanalysis Howard M. Heys and Stafford E.
More information0 Sets and Induction. Sets
0 Sets and Induction Sets A set is an unordered collection of objects, called elements or members of the set. A set is said to contain its elements. We write a A to denote that a is an element of the set
More informationConstructing Vectorial Boolean Functions with High Algebraic Immunity Based on Group Decomposition
Constructing Vectorial Boolean Functions with High Algebraic Immunity Based on Group Decomposition Yu Lou 1, Huiting Han 1, Chunming Tang 1, and Maozhi Xu 1,2 1 LMAM, School of Mathematical Sciences, Peing
More informationLinear Algebra. Paul Yiu. Department of Mathematics Florida Atlantic University. Fall 2011
Linear Algebra Paul Yiu Department of Mathematics Florida Atlantic University Fall 2011 Linear Algebra Paul Yiu Department of Mathematics Florida Atlantic University Fall 2011 1A: Vector spaces Fields
More informationA construction of Boolean functions with good cryptographic properties
A construction of Boolean functions with good cryptographic properties Jong H. Chung 1, Pantelimon Stănică 1, Chik-How Tan, and Qichun Wang 1 Department of Applied Mathematics, Naval Postgraduate School,
More informationOn a generalized combinatorial conjecture involving addition mod 2 k 1
On a generalized combinatorial conjecture involving addition mod k 1 Gérard Cohen Jean-Pierre Flori Tuesday 14 th February, 01 Abstract In this note, e give a simple proof of the combinatorial conjecture
More informationThe average dimension of the hull of cyclic codes
Discrete Applied Mathematics 128 (2003) 275 292 www.elsevier.com/locate/dam The average dimension of the hull of cyclic codes Gintaras Skersys Matematikos ir Informatikos Fakultetas, Vilniaus Universitetas,
More informationConstructions of digital nets using global function fields
ACTA ARITHMETICA 105.3 (2002) Constructions of digital nets using global function fields by Harald Niederreiter (Singapore) and Ferruh Özbudak (Ankara) 1. Introduction. The theory of (t, m, s)-nets and
More informationLinear Algebra (part 1) : Vector Spaces (by Evan Dummit, 2017, v. 1.07) 1.1 The Formal Denition of a Vector Space
Linear Algebra (part 1) : Vector Spaces (by Evan Dummit, 2017, v. 1.07) Contents 1 Vector Spaces 1 1.1 The Formal Denition of a Vector Space.................................. 1 1.2 Subspaces...................................................
More informationCharacterizations on Algebraic Immunity for Multi-Output Boolean Functions
Characterizations on Algebraic Immunity for Multi-Output Boolean Functions Xiao Zhong 1, and Mingsheng Wang 3 1. Institute of Software, Chinese Academy of Sciences, Beijing 100190, China. Graduate School
More informationMathematics Course 111: Algebra I Part I: Algebraic Structures, Sets and Permutations
Mathematics Course 111: Algebra I Part I: Algebraic Structures, Sets and Permutations D. R. Wilkins Academic Year 1996-7 1 Number Systems and Matrix Algebra Integers The whole numbers 0, ±1, ±2, ±3, ±4,...
More informationOptimal XOR based (2,n)-Visual Cryptography Schemes
Optimal XOR based (2,n)-Visual Cryptography Schemes Feng Liu and ChuanKun Wu State Key Laboratory Of Information Security, Institute of Software Chinese Academy of Sciences, Beijing 0090, China Email:
More informationA Questionable Distance-Regular Graph
A Questionable Distance-Regular Graph Rebecca Ross Abstract In this paper, we introduce distance-regular graphs and develop the intersection algebra for these graphs which is based upon its intersection
More informationA property concerning the Hadamard powers of inverse M-matrices
Linear Algebra and its Applications 381 (2004 53 60 www.elsevier.com/locate/laa A property concerning the Hadamard powers of inverse M-matrices Shencan Chen Department of Mathematics, Fuzhou University,
More informationAutomorphism group of the set of all bent functions
Automorphism group of the set of all bent functions Natalia N. Tokareva 1 Sobolev Institute of Mathematics, Novosibirsk State University, Novosibirsk, Russian Federation e-mail: tokareva@math.nsc.ru Abstract.
More informationCoins with arbitrary weights. Abstract. Given a set of m coins out of a collection of coins of k unknown distinct weights, we wish to
Coins with arbitrary weights Noga Alon Dmitry N. Kozlov y Abstract Given a set of m coins out of a collection of coins of k unknown distinct weights, we wish to decide if all the m given coins have the
More informationOn values of vectorial Boolean functions and related problems in APN functions
On values of vectorial Boolean functions and related problems in APN functions George Shushuev Sobolev Institute of Mathematics, Novosibirsk, Russia Novosibirsk State University, Novosibirsk, Russia E-mail:
More informationMin-Rank Conjecture for Log-Depth Circuits
Min-Rank Conjecture for Log-Depth Circuits Stasys Jukna a,,1, Georg Schnitger b,1 a Institute of Mathematics and Computer Science, Akademijos 4, LT-80663 Vilnius, Lithuania b University of Frankfurt, Institut
More informationChapter 1 Vector Spaces
Chapter 1 Vector Spaces Per-Olof Persson persson@berkeley.edu Department of Mathematics University of California, Berkeley Math 110 Linear Algebra Vector Spaces Definition A vector space V over a field
More informationA Hilton-Milner-type theorem and an intersection conjecture for signed sets
A Hilton-Milner-type theorem and an intersection conjecture for signed sets Peter Borg Department of Mathematics, University of Malta Msida MSD 2080, Malta p.borg.02@cantab.net Abstract A family A of sets
More informationLinear Codes, Target Function Classes, and Network Computing Capacity
Linear Codes, Target Function Classes, and Network Computing Capacity Rathinakumar Appuswamy, Massimo Franceschetti, Nikhil Karamchandani, and Kenneth Zeger IEEE Transactions on Information Theory Submitted:
More informationMaiorana-McFarland class: Degree optimization and algebraic properties
Downloaded from orbitdtudk on: Jan 10, 2019 Maiorana-McFarland class: Degree optimization and algebraic properties Pasalic, Enes Published in: I E E E Transactions on Information Theory Link to article,
More information